Soluble chitosan and poly-L-lysine are readily hydrolysed using lysozyme or chitosanase for chitosan, and trypsin, chymotrypsin or proteinase K for poly-L-lysine. For similar amounts of enzyme, chitosanase hydrolysed 57% of the chitosan, compared to 35% for lysozyme. In the case of poly-L-lysine, chymotrypsin and trypsin exhibited similar activities, hydrolysing approximately 41% of the polymer compared to proteinase K at only 16%. In contrast, chitosan and poly-L-lysine membranes, coating alginate beads, were almost totally inert to the respective hydrolytic enzymes. Less than 2% of the membrane weight was hydrolysed. It appears that either membrane material would be stable for in vivo application, and in particular in the protection of DNA during gastrointestinal transit. At chitosanase concentrations of 1.4 mg/ml and in the presence of sodium ions, 20% of the total double-stranded DNA was released from chitosan coated beads. An exchange of calcium for sodium within the bead liquefied the alginate core releasing DNA. The presence of calcium stabilized the alginate bead, retaining all the DNA. Highly pure DNA was recovered from beads through mechanical membrane disruption, core liquefaction in citrate and use of DNA spin-columns to separate DNA/alginate mixtures in a citrate buffer. DNA recovery efficiencies as high as 94% were achieved when the initial alginate/DNA weight ratio was 1000.